Theranos Wasn't Built on a Blood Test. It Was Built on a Share Class.

The most important thing Elizabeth Holmes ever invented was not a blood test. It was a share class. At Theranos, she held 100 votes for every single vote any other shareholder could cast, and she kept total control over who sat on the board.⁵ That arrangement raised no headlines and ran no commercials. It was the quiet machinery underneath everything else — and it is the reason a company that booked little more than $100,000 in real operating revenue could tell investors it had done $108 million, and have nobody inside the room positioned to say otherwise.²

The official story is that Theranos was a visionary technology startup undone when its science failed. The truer story is that the science was never the load-bearing pillar. The control structure was. Holmes didn't build a company that happened to have unusual governance — she built a governance fortress that happened to have a company inside it.

State the thesis plainly: Holmes's ownership structure was not a footnote to the strategy. It was the strategy. The 100-votes-per-share design, concentrating control of the board in one person, didn't just let her run the company her way — it methodically dismantled every accountability mechanism that might have surfaced the fraud years earlier. The same instrument founders use to protect a vision from short-term pressure was repurposed to protect a fiction from scrutiny.

A board built to impress, not to check

Look at who sat on the Theranos board and you understand the design. It was stacked with former Secretaries of State, former Defense Secretaries, former senators, and senior military officers — a wall of credentials that made the company look unimpeachable to anyone glancing in from outside.⁶ What that board conspicuously lacked was anyone who could evaluate a blood-diagnostics claim. No clinical-lab expertise. No relevant regulatory or scientific depth to challenge the central thing the company existed to do.⁶ The board was assembled for its photograph, not its judgment.

And because Holmes controlled board composition outright, the board was never a check on her — it was an extension of her. Reporting on the company's investigation indicates that any director who voiced a dissenting view was asked to resign.⁶ That is the whole mechanism in one sentence: in a normal company, a board exists so the CEO can be wrong and survive being corrected. At Theranos, the structure meant the CEO could not be corrected at all. The accountability loop was severed at the source.

She picked the investors who wouldn't ask

Control of the boardroom handled internal scrutiny. The funding strategy handled the external kind. The detail that gives the game away is who Theranos did not raise from. Many of the most experienced venture firms — the ones whose entire job is to demand proof a technology works — declined to invest, because Theranos refused to share that proof and insisted on secrecy.⁷ So the capital came from elsewhere. The company raised more than $700 million, and the round that cemented its roughly $9 billion paper valuation came largely from private equity rather than domain-expert venture capitalists.⁷¹

Read together, those two facts are a pattern, not a coincidence. The people best equipped to test the claims were structurally excluded — from the board by Holmes's control, and from the cap table by their own demand for evidence she would not give. What was left was a capital base willing to buy the narrative, and a governance design that guaranteed nobody on the inside could puncture it. Secrecy wasn't a personality quirk. It was a screen that selected for the investors who wouldn't ask the questions.

The control that made her a billionaire on paper, and worth zero in fact

Here is the part that turns the structure from clever into self-defeating. The thing that gave Holmes her power — voting control — was bought at the price of where she stood in the money. She held common stock, not preferred.³ On the way up, that didn't matter: the $9 billion paper valuation made her stake look like billions, and the press crowned her a self-made billionaire.⁴ But common stock sits at the back of the line. The SEC noted that, accounting for the liquidation preference, more than $750 million would have to be returned to defrauded investors and other preferred shareholders before Holmes could profit from her ownership at all.³ When Forbes marked the company down to roughly $800 million in 2016, the math did the rest — her stake collapsed to essentially nothing.⁴

So the same instrument did both jobs. The control structure let her wall off scrutiny long enough to inflate a $100,000 business into a $9 billion story.²⁷ And the kind of stock she held to keep that control meant the story was the only asset she ever truly owned. The moment the story died, so did the value. She was never standing on $4.5 billion — the Forbes figure that briefly made her famous.⁹ She was standing on the absence of anyone allowed to say there was nothing underneath.

Isn't this just normal founder control, gone wrong?

The fair objection is that founder control is ordinary and often good. Plenty of admired companies give founders super-voting shares precisely so they can ignore quarterly noise and build for the long run — and we don't call that fraud. By that logic, Theranos just had a bad founder, and the structure is innocent. That objection has real force, and it deserves a straight answer.

Two things separate Theranos from a healthy version of the same arrangement. First, degree: this went well beyond standard dual-class design, ultimately concentrating a supermajority of voting rights — the SEC's own enforcement leadership would later describe it as "nearly complete control" — in one person.¹⁰ A founder with strong control still answers to a real board and real co-owners; one with 99% of the votes and a hand-picked board of non-experts answers to no one. Second, the structure didn't merely permit the fraud — it was what made it survivable for years. The honest counter is right that control is neutral on its own. But the more total the control, the more it removes the very people whose job is to catch the lie. The settlement itself names the cure: the SEC required Holmes to convert her super-voting Class B shares to Class A and give up voting control entirely.¹ The remedy for the fraud was, precisely, dismantling the ownership structure that enabled it.

The conviction came later — wire fraud, four counts, in early 2022.⁸ But the verdict that explains Theranos was written years earlier, into the cap table. Holmes engineered a company where she could not be questioned, then filled it with claims that could not survive a question. The structure was the strategy, and the strategy worked exactly as designed — right up until the only check it could not wall off, the public record, finally arrived. Theranos didn't fail because the science was false. It failed because, eventually, someone with no super-voting shares got to ask.